Nabble - Subversion

[Issue 3154] Creating a branch from sparse working copy

2009-12-11T11:01:49Z

http://subversion.tigris.org/issues/show_bug.cgi?id=3154

------- Additional comments from tomtrias@... Fri Dec 11 11:01:48 -0800 2009 -------
This issue and a vote in favor of adding functionality to svn copy was mentioned
in the thread "Copy of a partial tree?" started by Kylo Ginsberg in svn users.
His final message of the matter can be found here:

http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=877133

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=463&dsMessageId=2429846

To unsubscribe from this discussion, e-mail: [issues-unsubscribe@...].

From forum: subversion - issues

[Issue 3154] Creating a branch from sparse working copy

2009-12-11T10:55:26Z

http://subversion.tigris.org/issues/show_bug.cgi?id=3154

------- Additional comments from tomtrias@... Fri Dec 11 10:55:26 -0800 2009 -------
Note that copy --parents does not preserve history from trunk itself (e.g. property changes) onto branches/b1, nor do we get copyfrom information for b1 in the log.

We are writing a system in which we cherry pick versions of files and directories to copy (instead of forcing copies of entire revisions) because it is more like the workflow our developers are used to and it
handles cases in a non-ideal world (like ours :-). We are using a .NET version of SVNMUCC that we hope to get re-incorporated into SharpSVN. Since we are letting developers pick and choose changes (including
important property changes on directories), we thought it would be ideal if we could copy a directory with depth empty and then copy in additional selected descendants.

Once that dream was dashed against the rocks of reality, we opted for the only solution available: a copy of the directory, followed by a delete of all of the children, followed by subsequent copies of
descendants (which may be copied from different revisions than that of the directory). Lather, rinse, repeat.

This works well enough, although it always shows the descendants as replaced, even if the conceptual application was a first time add with history. It also involves sending (many) additional delete commands
over the wire, when all we really wanted was a copy with limited depth.

The only real issues we see with adding depth to copy are:

1) Another flag or an additional depth enumeration ("WC"?) would be necessary for a WC-WC or WC-URL copy in order to indicate "use the depth as currently specified in the WC".

2) As it stands, the SVN log (and thus SVN users) would not differentiate between a fully recursive copy and a sparse copy.

As far as #2 is concerned, we already deal with that now to some extent. Since we are allowing cherry-picked changes, we want to display all of the changed paths associated with a revision. For our purposes,
we also want to explicitly show the implied additions and deletions from the copy or delete of a directory. To get an accurate picture of what actually went on in the revision, the log provides insufficient
information. Instead, we couple log information (in order to get the revprops) with diff --summarize information (we could use ls as well, but diff provides even more detail and we already have the author and
date from the log). Note that this information would also allow one to differentiate between various copy depths based upon what was actually copied. We could kill two birds with one stone if an additional
option were added to svn log to have it return all affected paths with copyfrom information (or at least an indicator regarding an explicit vs. implicit operation if separate copyfrom information is considered
overkill). If the paths also had separate associated information for text vs. property operations on each path (as svn diff does), so much the better...

Thanks.

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=463&dsMessageId=2429845

To unsubscribe from this discussion, e-mail: [issues-unsubscribe@...].

From forum: subversion - issues

[Issue 3550] New - Investigate streamy implementation of svn_fs_paths_changed2()

2009-12-11T09:21:33Z

http://subversion.tigris.org/issues/show_bug.cgi?id=3550
Issue #|3550
Summary|Investigate streamy implementation of svn_fs_paths_cha
|nged2()
Component|subversion
Version|1.6.x
Platform|All
URL|
OS/Version|All
Status|NEW
Status whiteboard|
Keywords|
Resolution|
Issue type|ENHANCEMENT
Priority|P3
Subcomponent|libsvn_fs
Assigned to|cmpilato
Reported by|cmpilato

------- Additional comments from cmpilato@... Fri Dec 11 09:21:32 -0800 2009 -------
The current implementation of svn_fs_paths_changed2() isn't streamy -- you give
it a root, and it gives you a potentially massive hash back. For scalability
reasons, we should consider developing a streamy version of that interface.

BONUS: Ensure that callers of this function never get an svn_fs_path_change2_t
item with an unknown node kind.

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=463&dsMessageId=2429821

To unsubscribe from this discussion, e-mail: [issues-unsubscribe@...].

From forum: subversion - issues

Re: Feature request: compare file to other working copy from context menu

2009-12-11T07:49:03Z

Hello Matthias,

Most likely the simplest solution would be to add both working copies to the
same SmartSVN project. Then you can select both files and invoke Query|Compare
2 Files.

--
Best regards,
Thomas Singer
=============
syntevo GmbH
http://www.syntevo.com
http://blog.syntevo.com

mweyh <m.weyh@...> wrote:

>
> Hello,
>
> I have a lot of working copies of different branches or tags of the same
> repository. Most of the time I fix an issue in one of the branches and I
> need to merge that change to others before actually committing the changes.
> It would be very helpful if there was a way of comparing a file from one
> working copy with another.
>
> Maybe this could be included in the "compare with" dialog as an additional
> entry "other working copy" with a dropdown box of the available SmartSVN
> projects. If a file would not exist in that other working copy SmartSVN
> could just issue an error.
>
> best regards,
> Matthias
> --
> View this message in context:
> http://old.nabble.com/Feature-request%3A-compare-file-to-other-working-copy-from-context-menu-tp26742654p26742654.html
> Sent from the SmartSVN mailing list archive at Nabble.com.
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/smartsvn/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/smartsvn/join
(Yahoo! ID required)

<*> To change settings via email:
smartsvn-digest@...
smartsvn-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
smartsvn-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/

From forum: SmartSVN

RE: Migration to SVN

2009-12-11T07:29:11Z

>

Linedata Services (UK) Ltd
Registered Office: Bishopsgate Court, 4-12 Norton Folgate, London, E1 6DB
Registered in England and Wales No 3027851 VAT Reg No 778499447

-----Original Message-----

> From: barveaniketan@... [mailto:barveaniketan@...]
> Sent: 11 December 2009 13:20
> To: users@...
> Subject: Re: Migration to SVN
>
> Hi All,
>
> I am also working on a big project to migrate our source code
> from an old system (CMS running on VMS OS) to Subversion.
>
> So could you please tell me what process you have followed
> for CMS to SVN MIGRATION?
>
> I want know below mentioned points.
> 1. How CMS data is exported to local directory. In my case I
> have around 1000 elements do take copy of each element
> manually its very time consuming & critical job.

We had perl scripts to work with CMS so I never knew the direct CMS commands. However, we worked with classes which I think *are* CMS specific. How to use them in your case it much depends on how you have organised your code I think.

> 2. Suppose once you have exported data to local directory.
> Then what is the procedure to export data to the subversion.

You can use a series of "svn import" or maybe just one and then some "svn copy" and "svn commit". Again it depends on how your code is organised and how you want it organised in Subversion.

Giulio

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2429802

Please start new threads on the <users@...> mailing list.
To subscribe to the new list, send an empty e-mail to <users-subscribe@...>.

Nabble - Subversion

[Issue 3154] Creating a branch from sparse working copy

[Issue 3154] Creating a branch from sparse working copy

[Issue 3550] New - Investigate streamy implementation of svn_fs_paths_changed2()

Re: Feature request: compare file to other working copy from context menu

RE: Migration to SVN

Re: Query: svn externals and export

Re: Password problems on OpenVMS

Re: Password problems on OpenVMS

RE: Re: Query: svn externals and export

Re: Migration to SVN

Re: Subversion - Llocked file could not able to commit the changes

RE: Subversion - Llocked file could not able to commit the changes

Re: stash (was Re: revert --force)

[Issue 3549] New - New option --remove-unversioned-files to cleanup

SVN API in Perl

Feature request: compare file to other working copy from context menu

Re: stash (was Re: revert --force)

Re: stash (was Re: revert --force)

Re: svnadmin: Decompression of svndiff data failed - how to recover?

Re: Subversion Install Help

Re: Error starting apache 2 after loading mod_dav_svn when upgrading to subversion 1.6.6

Re: apache crashes with mod_dav_svn

RE: Query: svn externals and export

Re: Query: svn externals and export

RE: Re: Query: svn externals and export

Re: Feature request: Add an "Automatically lock file before opening for edit" option

Re: Suggestion: log --same-path

RE: revert --force

Re: stash (was Re: revert --force)

Subversion with SASL on Windows XP

Re: revert --force

Re: About SVN

About SVN

Re: revert --force

[Issue 3548] Track dav_error API changes in mod_dav_svn (so we can compile against Apache 2.4)